Volume 134, Issue 6, Pages 933-944 (September 2008) Identification of a Lipokine, a Lipid Hormone Linking Adipose Tissue to Systemic Metabolism  Haiming Cao, Kristin Gerhold, Jared R. Mayers, Michelle M. Wiest, Steven M. Watkins, Gökhan S. Hotamisligil  Cell  Volume 134, Issue 6, Pages 933-944 (September 2008) DOI: 10.1016/j.cell.2008.07.048 Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 1 Whole-Body Lipid Profiling (A) Dendrograms of hierarchical clustering of subjects, based on lipid profiles. Only fatty acids with a p value < 0.05 from a one-way ANOVA were included. R, regular diet; HF, high-fat diet; KO, FABP−/− mice. (B) Scatter plots of principal components from principal component analysis on subjects. Only fatty acids with a p value < 0.05 from a one-way ANOVA were included. (C) Strength of differences between diets in the WT and FABP−/− (KO) mice over all fatty acids in all lipid classes. Each point on a line indicates the number of p values of one group that are smaller than the p value of equal ranking of the other group. Cell 2008 134, 933-944DOI: (10.1016/j.cell.2008.07.048) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 2 Adipose De Novo Lipogenesis and Palmitoleate Production (A) Difference of major fatty acids in diacylglycerol, free fatty acid, phospholipid, and triglyceride fractions in adipose tissue between the means of WT and FABP−/− mice. (B) Lipid class composition analysis for fatty acids in adipose tissue. The top 36 metabolites are shown. The F statistics from a one-way ANOVA are displayed as red diamonds over the distribution of F statistics from permuted data. The black line is the maximum F statistic observed in 100 permutations. The heat map displays the observed data, centered to the mean of the control group and scaled by the standard deviation of all observations. (C) Plasma fatty acids in WT or FABP−/− (KO) mice. (D) Percentile suppression of palmitoleate by HFD in adipose tissue of WT or KO mice. Error bars represent the SEM. Cell 2008 134, 933-944DOI: (10.1016/j.cell.2008.07.048) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 3 Metabolic Regulation by Plasma Lipids (A) SCD-1 promoter activities in control hepatocytes (No lipid) or hepatocytes treated with plasma lipids extracted from WT or FABP−/− (KO) mice. (B) SCD-1 promoter activities in hepatocytes treated with fatty acid mixtures resembling the ratio of plasma fatty acids from WT mice (WT Mix), KO mice (KO Mix), KO mice with the concentration of palmitoleate of WT mice (KO Mix, − 16:1n7), or KO mice with the concentration of palmitoleate, palmitate, and stearate of WT mice (KO Mix, − 16:1n7, + 16:0 18:0). Asterisk, significantly different from WT mix; a and b, significantly different from KO Mix. (C) Insulin-stimulated AKT phosphorylation in C2C12 myotubes treated with plasma lipids extracted from WT or KO mice. Values of bar plot were determined by phospho-AKT ELISA, and corresponding immunoblotting results are shown as insets. (D) MCP1 in conditional medium of adipose explants treated with plasma lipids of WT or KO mice. (E) SCD-1 promoter activities in hepatocytes treated with fatty acids. All fatty acids were used at 300 μM final concentration, except 16:1n7 2x is 600 μM. AA, arachidonic acid; asterisk, significantly different from controls; a, significantly different from AA-treated cells; b, significantly different from palmitate-treated cells. (F) Immunobloting of Flag-tagged SCD-1 in hepatocytes treated with control, palmitate, and palmitoleate. Tubulin was used as loading control. (G) AKT phosphorylation in C2C12 myotubes treated with fatty acids and insulin. a, significantly different from cells treated with insulin; b, significantly different from cells treated with both insulin and palmitate. The bottom panel shows the corresponding immunoblotting of total and phosphorylated AKT in cells treated with pooled lipids. (H) Glucose uptake in C2C12 myotubes treated with insulin or palmitoleate. The bottom panel is immunoblotting of C2C12 cell lysates treated with insulin or palmitoleate with anti-Glut1 or Glut4 antibodies. Asterisk, p < 0.05. Error bars represent the SEM. Cell 2008 134, 933-944DOI: (10.1016/j.cell.2008.07.048) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 4 Systemic Palmitoleate Metabolism (A) Free fatty acid and triglyceride flux among adipose, muscle, and liver tissues. Means and SEM of palmitoleate in free fatty acids (FFA) and triglyceride (TG) in plasma and adipose, muscle, and liver tissues of WT and FABP−/− mice are shown. The predicted lipid flux based on the tissue lipid profile pattern is indicated by the direction of arrows. Muscle fatty acids are mainly derived from liver in the form of VLDL-associated TG and from adipose tissue in the form of FFAs. (B) Percentile suppression of palmitoleate by HFD in TG fraction of muscle tissue in WT or KO mice. (C) Total palmitoleate in liver of WT or KO mice. (D) Triglycerides in liver of KO mice injected with control or SCD-1 adenoviruses. Asterisk, p < 0.05. Error bars represent the SEM. Cell 2008 134, 933-944DOI: (10.1016/j.cell.2008.07.048) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 5 Lipid-Mediated Regulation of SCD-1 Promoter Activity In Vivo (A) Schematics of SCD-1 −1500bp promoter. PUF, polyunsaturated fatty acid response element. (B) Regulation of WT and mutant SCD-1 promoter activities by fatty acids. AA, arachidonic acid. (C) Regulation of wild-type and mutant SCD-1 promoter activities by plasma lipids. Plasma lipids extracted from WT or FABP−/− (KO) mice were used to treat FAO cells infected with WT or mutant SCD-1 promoter-carrying adenoviruses, and luciferase assays were performed as described in the Experimental Procedures. (D) In vivo regulation of WT and mutant SCD-1 promoter activities. Luciferase assays were performed on liver tissues of WT or KO mice injected with WT or mutant SCD-1 reporter-carrying adenoviruses. (E) Liver gene expression in mice infused with vehicle, TG:palmitate or TG:palmitoleate. Asterisk, p < 0.05. Error bars represent the SEM. Cell 2008 134, 933-944DOI: (10.1016/j.cell.2008.07.048) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 6 Regulation of Lipogenic Genes in Adipose and Liver of WT and FABP−/− Mice (A) Lipogenic gene expression in adipose tissue of WT and FABP−/− (KO) mice. Gene expression patterns in epididymal fat pad were determined by quantitative real-time PCR. (B) Regulation of lipogenic gene promoter activites by FABPs in differentiated adipocytes. Differentiated FABP−/− adipocytes that expressed SCD-1 or FAS promoter luciferase constructs were infected with GFP, aP2, or both aP2 and mal1 adenoviruses, and luciferase activities were determined by dual-glow luciferase system. a, significantly different from GFP-adenovirus-infected cells; b, significantly different from aP2-adenovirus-infected cells. (C) Regulation of lipogenic gene expression in adipose tissue of mice treated with aP2 inhibitor. Gene expressions in epididymal fat pads from mice treated with vehicle or aP2 inhibitor were determined with quantitative real-time PCR. (D) Differential regulation of lipogenic gene expression in adipose and liver tissues. Levels of the mRNAs in liver and epididymal fat pad from mice fed with HFD were determined, and WT liver and KO adipose were set as 1. ACC1, acetyl-CoA carboxylase1; DGAT1, acyl-CoA:diacylglycerol acyltransferase 1. Asterisk, p < 0.05. Error bars represent the SEM. Cell 2008 134, 933-944DOI: (10.1016/j.cell.2008.07.048) Copyright © 2008 Elsevier Inc. Terms and Conditions

Figure 7 Regulation of Insulin Signaling and Glucose Metabolism by Palmitoleate (A) Basal and insulin-stimulated phosphorylation of insulin receptor (IR), insulin receptor substrate-1 (IRS-1) and -2 (IRS-2), and AKT in liver of mice infused with vehicle (Veh), TG:palmitoleate (16:1n7), or TG:palmitate (16:0). Representative blots are shown, and quantifications on the right are averaged results of three mice in each treatment group. (B) Basal and insulin-stimulated phosphorylation of insulin receptor (IR), insulin receptor substrate-1 (IRS-1), AKT, and GSK in muscle tissues of mice infused with vehicle (Veh), TG:palmitoleate (16:1n7) or TG:palmitate (16:0). Representative blots are shown, and quantifications on the right are averaged results of three mice in each treatment group. (C) Basal hepatic glucose production (bHGP) of mice infused with vehicle (Veh) or palmitoleate (16:1n7). (D) Glucose infusion rate (GIR) of mice treated with vehicle (Veh) or palmitoleate (16:1n7) during hyperinsulinemic-euglycemic clamp. (E) Hepatic glucose production (Clamp HGP) of mice infused with vehicle (Veh) or palmitoleate (16:1n7) during hyperinsulinemic-euglycemic clamp. (F) Glucose disposal rate (RD) in mice infused with vehicle (Veh) or palmitoleate (16:1n7) during hyperinsulinemic-euglycemic clamp. (G) Regulation of systemic metabolic responses by adipose-derived lipid hormones (Lipokines). In parallel with a variety of adipokines, specific lipids released from adipocytes in response to physiological stimuli act at remote sites including liver and muscle and regulate systemic lipid and carbohydrate metabolism. Lipid chaperones negatively regulate one of these lipokines, C16:1n7-palmitoleate, and in the absence of these FABPs, strong flux of palmitoleate from adipose tissue to liver and muscle results in improved metabolic responses. Error bars represent the SEM. Cell 2008 134, 933-944DOI: (10.1016/j.cell.2008.07.048) Copyright © 2008 Elsevier Inc. Terms and Conditions